Figure 1.
Schematic representation of experimental design.
During adaptation, animals were exposed to six CS− only and the entire session was repeated six hours later (Day 1). Conditioning took place on the following day (Day 2): the CS+ was presented three times, co-terminated with an electric footshock, during each conditioning session. Conditioned fear memory was tested on day 3 ( = extinction training); six consecutive extinction training sessions were carried out (R1 to R6), with 30 minutes intervals between sessions. Extinction memory was recalled on day 4 in one session (E). All sessions on day 3 and 4 were identical, and contained four CS− and four CS+ presentations. Stimulus parameters and intervals are marked.
Figure 2.
Freezing behavior and theta synchronization in naive 5-HTT mice at different stages of fear extinction.
A) Freezing behavior during the first CS+ across R1-R6 and E in all three genotypes. Freezing in all the three genotypes declined from R1-R6, indicating the standard extinction of fear memory. Twenty four hours later, when all the three genotypes were tested for fear extinction recall, 5-HTT−/− mice showed significant impaired extinction recall (E), represented by significantly higher levels of freezing compared to 5-HTT+/− and 5-HTT+/+ mice (p<0.001). B) Stimulus-related theta correlation in 5-HTT mice at different stages of fear extinction. Averaged cross-correlation during the presentation of the first CS+ between LA and IL of mPFC during extinction (R1-R6) and extinction recall (E) in 5-HTT naive mice is shown. When compared with 5-HTT+/+, 5-HTT−/− mice showed significantly increased cross-correlation during extinction training (note significant at R4 and R6; p<0.05). In 5-HTT−/− mice, the correlation values remained high across the extinction training and being significantly higher than for 5-HTT+/+ mice during E. C) Freezing-related averaged cross-correlation during the first CS+ between LA and IL of mPFC during extinction training (R1-R6) and extinction recall (E) in 5-HTT naive mice. Note, that the 5-HTT−/− mice showed significantly increased cross-correlation during extinction recall (p<0.05, compared to 5-HTT+/− and 5-HTT+/+− mice). Values are mean ± SEM; *p<0.05, #p<0.001.
Figure 3.
Theta synchronization in different stages of fear extinction in 5-HTT (+/+ and −/−) (A) control and (B) socially defeated (loser) mice.
Local field potentials (LFPs) from LA, and PFC were simultaneously recorded during early retrieval (R1) and extinction recall while monitoring the animals' behavior. The figure shows representative LFP traces (20 s) for each recording area from R1 and E during the first CS+ presentation for a control and a loser 5-HTT+/+ and −/− mouse, respectively. The behavior of the mice at the particular time is indicated above the traces (f, freezing; r, risk assessment; e, exploration; g, grooming; q, quiet), and wavelet transforms of the 20 s segments for each recording area are displayed underneath. The (a) stimulus- and (b) freezing-related cross-correlations are shown beneath the LFP traces, indicating theta synchronization. Independent of the genotype, both the control and socially defeated mice showed increased freezing behavior accompanied by high stimulus- and freezing-related theta synchrony at R1. Note, during recall of extinction (E), the 5-HTT−/− control and loser mice showed increased freezing behavior and increased theta synchrony, whereas the 5-HTT+/+ control and loser mice showed low freezing behavior and low theta synchronization between LA and PFC.
Figure 4.
Freezing behavior and theta synchronization in 5-HTT control and loser mice at different stages of fear extinction.
A) Averaged freezing behavior, D) stimulus-related and G) freezing-related LA-mPFC theta correlation during the first CS+ across R1-R6 and E in 5-HTT+/+ control and loser mice. 5-HTT+/+ loser mice showed a slightly delayed extinction learning (significant at R4) and impaired recall of extinction (significant at E) accompanied by increased stimulus-related LA-mPFC synchronization during extinction training (significant at R4). Compared to controls, 5-HTT+/+ loser mice showed increased freezing-related theta synchronization at early retrieval sessions (significant at R2). B) Averaged freezing behavior, E) stimulus-related and H) freezing-related LA-mPFC theta correlation during the first CS+ across R1-R6 and E in 5-HTT+/− control and loser mice. 5-HTT+/− loser mice showed a significantly delayed extinction learning (R3, R4) and impaired recall of extinction (E) accompanied by significant increased stimulus-related LA-mPFC synchronization across extinction learning and recall of extinction. Note, compared to controls, 5-HTT+/− loser mice showed a highly significant increase of freezing-related theta synchronization across extinction learning (R1-R6) and during recall of extinction (E). C) Averaged freezing behavior, F) stimulus-related and I) freezing-related LA-mPFC theta correlation during the first CS+ across R1-R6 and E in 5-HTT−/− control and loser mice. 5-HTT−/− loser mice showed a significant delayed extinction learning (R4, R5). Impaired recall of extinction is indicated by high freezing level in both groups. The stimulus-related correlations in 5-HTT−/− loser mice were not significantly different from 5-HTT−/− controls, but both the groups showed high synchronized LA-mPFC activity across R1-R6 and at E. Periods of freezing are accompanied by high freezing-related theta synchronization during extinction learning in 5 HTT−/− control (R1-R3) and 5 HTT−/− loser mice (R1-R6) and during extinction recall in both groups. Values are mean ± SEM; *p<0.05, **p<0.01, #p<0.001.
Figure 5.
Freezing behavior and theta synchronization in 5-HTT loser mice (all genotypes) at different stages of fear extinction.
A) Freezing behavior during the first CS+ across R1–R6 and E. Freezing in all three genotypes declined from R1-R6, indicating the standard extinction of fear memory, but both 5-HTT+/− and −/− mice showed delayed extinction (in −/−, significant at R4, R5 and in +/− significant at R5). Twenty four hours later, when all three genotypes were tested for fear extinction recall, 5-HTT−/− mice showed significantly higher freezing behavior (impaired recall) than 5-HTT+/+ and +/− mice, and 5-HTT +/− were significantly different from 5-HTT+/+ loser mice. B) Stimulus-related theta synchronization in 5-HTT loser mice at different stages of fear extinction. Compared with 5-HTT+/+ (significant at R5) and 5-HTT−/− losers (significant at R3 and R5), the 5-HTT+/− loser mice showed a significantly higher correlation during extinction. 5-HTT +/− and −/− loser mice displayed significantly higher correlations at extinction recall (E) compared to 5-HTT+/+. C) Freezing-related averaged cross-correlation during the first CS+ between LA and IL of mPFC during extinction training (R1-R6) and extinction recall (E) in 5-HTT loser mice. Note high freezing-related theta synchronization in all three genotypes. Values are mean ± SEM; *p<0.05, #p<0.001 compared to 5-HTT+/+; $p<0.05 compared to 5-HTT+/−.
Figure 6.
Directionality and time-lag of local field potentials between LA and mPFC at different stages of fear extinction.
Negative values indicate mPFC leading, positive values indicate LA leading. A) Time-lag between LA and mPFC in 5-HTT+/+ naive control and loser mice. In controls, the interaction is started with LA (R1, R2), but later directionality between both is driven by the mPFC (R3, R4, R5 and E), and the amygdala-prefrontal cortex interactions are significantly different between R1 and E (Unpaired t-test, * p<0.05). In 5-HTT+/+ loser mice, mPFC is dominating across the entire extinction training, but the directionality changes into LA dominance during E and is significantly different in E between control and loser mice (Unpaired t-test, § p<0.05). B) In both 5-HTT +/− control and loser mice, the mPFC is dominating across R1-R6 and at E. C) In both 5-HTT −/− control and loser mice, the mPFC is leading across R1-R6, but LA is leading in 5-HTT −/− loser mice at E. Values are mean ± SEM.